chg_mag: Change Magnitude Image for CVAPS

Description Usage Arguments Details Value References Examples

View source: R/chg_mag.R

Description

This code calculate the change magnitude image for the Change Vector Analysis in Posterior Probability Space (CVAPS) method of Chen et al. 2011. Use the change magnitude image and use it in conjunction with the change direction image from chg_dir to map areas of change and no-change. The threshold can be determined using DFPS (to use the Double Window Flexible Pace Search method, from Chen et al. 2003) or threshold (which uses an unsupervised method).

Usage

1
chg_mag(t1p, t2p, filename, overwrite = FALSE, ...)

Arguments

t1p

time 0 posterior probability Raster*

t2p

time 1 posterior probability Raster*

filename

(optional) filename for output change magnitude RasterLayer

overwrite

whether to overwrite existing files (otherwise an error will be raised)

...

additional parameters to pass to rasterEngine

Details

This function will run in parallel if a parallel backend is registered with foreach.

Value

Raster* object with change magnitude image

References

Chen, J., P. Gong, C. He, R. Pu, and P. Shi. 2003. Land-use/land-cover change detection using improved change-vector analysis. Photogrammetric Engineering and Remote Sensing 69:369-380.

Chen, J., X. Chen, X. Cui, and J. Chen. 2011. Change vector analysis in posterior probability space: a new method for land cover change detection. IEEE Geoscience and Remote Sensing Letters 8:317-321.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
## Not run: 
t0_train_data <- get_pixels(L5TSR_1986, L5TSR_1986_2001_training, "class_1986",training=.6)
t0_model <- train_classifier(t0_train_data)
t0_preds <- classify(L5TSR_1986, t0_model)
t1_train_data <- get_pixels(L5TSR_2001, L5TSR_1986_2001_training, "class_2001", training=.6)
t1_model <- train_classifier(t1_train_data)
t1_preds <- classify(L5TSR_2001, t1_model)
t0_t1_chgmag <- chg_mag(t0_preds$probs, t1_preds$probs)

## End(Not run)

azvoleff/teamlucc documentation built on June 4, 2017, 12:08 a.m.